Signaling Pathways in the Pathogenesis and Treatment of Diamond Blackfan Anemia

April 09, 2021

Information

Kathleen Sakamoto, MD, PhD
Shelagh Galligan Professor of Pediatrics, Stanford University
YCCEH Invited Speaker
April 8, 2021

ID6409

To CiteDCA Citation Guide

00:17Alright guys, why don't we go ahead
00:19and get started so we're having a
00:21zoom webinar format for this talk so
00:23Doctor Sakamoto will give her talk and
00:24then at the end we'll have questions.
00:26I think you're going to have to
00:28put your questions either in the
00:30Q&A or in the chat and then I can
00:32read them to her just given
00:34the format of this talk.
00:36So I'm Jeannie Hendrickson.
00:37I'm representing the
00:38Yellow Cooperative Center
00:40of Excellence in hematology
00:41for the enrichment program,
00:42and we're extremely happy to have
00:45Doctor Sakamoto who's giving a talk,
00:47a talk to us all the way from Stanford.
00:50Virtually, of course,
00:51she's a professor in pediatric
00:53hematology oncology at Stanford,
00:54and her research involves signaling
00:56pathways and gene regulation in
00:58normal and aberrant masterpieces,
01:00including bone marrow failure.
01:01And today she's going to focus her talk on
01:04this as it relates to Diamond Black fan.
01:07Nina, so I will turn it over
01:09to her and again thank you so
01:11much for joining us today.
01:14Thank you so much Gene.
01:15I really appreciate the invitation.
01:17I also want to thank Pat and Diane
01:19especially for this kind of rotation
01:21and what I'm going to do today for the
01:24next 45 minutes or so is to talk about
01:26some of the recent work that we've
01:28been doing on signaling pathways in
01:30the pathogenesis and treatment of TBI.
01:33So just to say I have no good disclosure,
01:35I have no disclosures that
01:36are related to this work.
01:38So Diamond life and anemia is a
01:40very rare boomer failure syndrome,
01:42and I'll discuss some of the background
01:45and clinical features of this disease.
01:48Also, the role of an email like
01:50high knees in DBA models and then
01:53targeting and OK for the treatment
01:56of DBA and the last part of the talk.
01:59I will talk about a new project on 71
02:02in erythropoiesis and EPA pathogenesis.
02:06So DPA is a very rare congenital
02:08bone marrow failure syndrome that's
02:10associated with macrocytic anemia,
02:13congenital defects, Anna risk of cancer.
02:15This disease is generally diagnosed in
02:18early childhood, less than a year of age.
02:21The incidence is seven in a million
02:24and approximately 20 to 40 new cases
02:27are diagnosed per year in the US,
02:30in Canada.
02:34Download iPad anemia.
02:36Clinical features are quite diverse and
02:39there are many patients who present
02:42with short stature shown here and
02:46these Twins upper limb abnormalities
02:48including thumb and facial and pallet
02:52malformations which can occur with certain
02:55ribosomal protein subunit mutations.
02:58Patients can present with small eyes,
03:00kidney defects, but only for 30% of
03:03all patients have physical findings,
03:05which makes it very challenging
03:08when you have a patient who may not
03:11manifest significant anemia and may
03:13be identified by their adult intern,
03:16internist, physician,
03:17who notices that they have microcytosis
03:19but no other physical features,
03:21and we've had a few of these patients
03:25come to clinic in our at Stanford.
03:29The treatment for Diamondback anemia
03:31is typically steroids, chronic ritzel,
03:33transfusions and stem cell transplantation.
03:36For patients who are steroid refractory
03:38or chronically transfusion dependent,
03:40and these are all associated
03:43with significant morbidities,
03:44including immunosuppression,
03:45iron overload, graph versus host disease,
03:48there are newer therapies that have
03:50come been proposed.
03:52For example, L, leucine, the amino acid.
03:55Many of you are familiar
03:57with the clinical trial.
03:59It's been directed by Jeff Lipton,
04:02an Adreno blocos.
04:03They showed a modest
04:05improvement with leucine,
04:06although the doses were
04:08much lower and for safety,
04:10but there were some patients who
04:13did experience improvement in
04:14the transfusion requirements.
04:16So Tatter septis Eligant wrapped that
04:19inhibits the TGF beta signaling pathways
04:21that also has been shown recently
04:24to improve with police in patients,
04:27for example with Milo dysplastic syndrome.
04:30But most of them are being
04:33cronian investigated.
04:37Why does summer deficiency in bone
04:39marrow failure syndrome is become
04:41more and more common in diseases
04:43such as Shockman Diamond syndrome
04:45which is another congenital bone
04:48marrow failure syndrome with other
04:50physical findings, deletion 5 Q.
04:52Milo dysplastic syndromes which
04:54is associated with our PS14
04:57haploinsufficiency that's been described
04:58by Ben Ebras group several years ago.
05:01These are all considered diseases
05:03that are ribosome opathy's with
05:05defective ribosome Biogenesis.
05:07And function.
05:08Over 80% of DBA patients have
05:11mutations in ribosomal protein.
05:13Some units resulting in ribosome dysfunction,
05:16an impaired protein translation.
05:19This is just a pie chart that shows
05:22the various percentages of patients
05:24who have ribosomal mutations.
05:26You can see that among the most
05:29common is RPS 1925% RP 11.
05:31About 5% in RP 11 and five have been
05:34associated more with craniofacial defects,
05:37and there are several other
05:39mutations have been identified but
05:4130% have non ribosomal mutations.
05:43In an example is what Vijay Sankar
05:46had reported that God would.
05:48Mutations can also contribute
05:49to this disease.
05:51But there are several more mutations
05:52that have yet to be identified.
05:56This is a schematic that basically
05:58summarizes some of the defects that
06:00have been identified in the ribosomes
06:03are made synthesis pathway primarily
06:05the Treacher Collins syndrome, which,
06:08Interestingly does not usually
06:10manifest with bone marrow findings.
06:12We have just keratosis congenita
06:14which most people are familiar
06:17with that can cause fibrosis as
06:19well as bone marrow failure.
06:21Cartilage hair hypoplasia has been
06:23associated with macrocytic anemia.
06:25And then of course, TBA,
06:27which can involve the small subunit
06:29or large subunit 60 S or 40th,
06:31and then shockman diamond syndrome.
06:36It's still very interesting that
06:38this disease, which is germline
06:40mutation of these Robertson,
06:42will proteins in other mutations
06:44involved primarily the erythroid
06:47lineages and what I thought is in
06:49this very simplified version of Rip
06:52Oasis is that the cells increase
06:54in number as they become Earth,
06:56will glass and somehow this results in
06:59increased requirement for ribosomes,
07:01most likely due to increase.
07:03Protein translation is required for
07:05cell proliferation and differentiation.
07:09The normal hematopoietic treated Mary.
07:11Are you familiar with already from?
07:13He might have put stem cells to
07:15come in my local gender Mega
07:17Carey Service Rd Ripper genders,
07:18and then the birth funding
07:20isn't calling from Eunice.
07:21Very earlier it'll bus stage all lead to
07:24eventually the maturing of red blood cells.
07:27In DBA, these very subunit mutations
07:30that have been described seemed
07:32to be a result in a block in
07:35early committed with blaster be.
07:37If you be a few istage,
07:40mostly resulting from haploinsufficiency
07:42inducing mutation and we know that
07:45homozygous mutations result in
07:47mouse models as well as in humans.
07:50Embryonic lethality so helpful
07:52insufficiency or heterozygous
07:53mutations are typically what results.
07:55What results in this disease.
07:58Ultimately leading to anemia.
08:01So we started out looking at
08:03variety of signaling molecules,
08:05one that was of interest to us initially
08:08was a transcription factor MYB and
08:10we know that this is a very important
08:14protein that regulates erythropoiesis.
08:16It's also associated with my little keemia.
08:19When Aberrantly expressed and it's been
08:22reported that one of the kinases that
08:25activates or phosphorylates Mibiz,
08:27NIMAL, Iconis,
08:28and this kinese is Assyrian
08:30threatening kinese.
08:31That is a revolutionary and evolutionary
08:34conserved margin activated protein
08:36kinase in the map kinase family member.
08:39Originally described when mutated
08:41result results in I development
08:44defects in just saffola is highly
08:46expressed in neural tissues and
08:49plays a critical role in a number of
08:52important cellular functions through
08:54the regulation of various transcription
08:57transcription factors such as nib.
09:00An OK also regulate signaling
09:02pathways involving wind beta catenin
09:04active in aisle 6 an notch,
09:05which are all very critical
09:08for normal hematopoiesis.
09:10So this is work done by Mark Walsh,
09:13who is a former postdoctoral
09:14fellow in my lab,
09:15and now is an instructor in Pediatrics.
09:19We used a model that had been used
09:21prior in my lab where we would
09:24transduced human CD 34 positive cells,
09:27hematopoietic stem progenitor
09:28cells with small hairpin RNAs,
09:30then knocked down either RPS 19 or RPO 11.
09:34Making lentiviral constructs with
09:36transduced and then show that we
09:39could get help low or half the
09:42expression at the protein level by
09:44this Western blot or M RNA levels.
09:47And this is the model that we used.
09:49And when Mark looked at the expression
09:52of an OK throughout, he matter.
09:54Police is we didn't see any
09:56significant differences in our bar.
09:58RPS 19 RP 11 knock down now.
10:01In contrast,
10:02looking at phosphorylation or activation
10:04of an OK when Mark noticed was that
10:08taking three different substrates,
10:09either an OK itself this is auto
10:13phosphorylates, nib or wrapped,
10:14or that there is significant increase
10:17in the activation of this kinese
10:20in the 1st 10 days of hematopoiesis
10:23and so this we decided to focus
10:26on an OK and looking at mid target
10:29genes with RPS 19 lockdown.
10:32Again, HSBC is,
10:33we see that their expression
10:34of Alamo two in California were
10:37significantly decreased.
10:38However,
10:38with the expression or knockdown of an OK,
10:41we see that that we could partially
10:43rescue the expression of these two genes.
10:46Obviously there other Trump,
10:47you know the regulators of nib,
10:49so that makes a lot of sense to us.
10:52I'm sorry, Kathy,
10:53can you go back one slide?
10:57What is one more? I'm sorry.
10:59What are you showing on the PC MIB?
11:03One forward PC MIB an?
11:05What's the PC MIB? There is
11:07middling. Phosphorylated
11:08yeah phosphorylated. MIB is
11:10that active? Is phosphorylated
11:11MIB active? Yes yes
11:13so. So that is a key phosphorylation
11:16site that's recognized by in OK. OK,
11:19so any OK is phosphorylating
11:21in activating them? Yes OK
11:24just clarifying yes and I'll talk
11:26more about that in a few slides.
11:29Yeah and so so now.
11:32How does an OK regulate nib actually what
11:35happens is that activated and OK then
11:38phosphorylates mid to ubiquitinated nib,
11:41an result in 26 S proteasome degradation.
11:44So if we have an OK and we knock it down.
11:49RPS 19 meter put extent progenitor cells.
11:52What we actually see is stabilization of MIB.
11:55So you see that the protein
11:57levels are higher and Konan OK.
11:59It's knocked down and this
12:01again this is in RPS.
12:0319 knockdown cells so you can see that
12:06if we again look at phosphorylated nip,
12:09that compared to when we knock down
12:11in OK we see higher levels of mid
12:14phosphorylation as we would expect
12:16because MIB results in degradation
12:18and OK phosphorylation results in.
12:20Amid degradation and also here only
12:23milk is ubiquitinated without an OK.
12:26So you see higher levels,
12:29so without an OK it's not degraded.
12:34So we also looked at C71 expression in
12:38this schematic schema that you see to
12:41indicate Orthop Oasis will Mark did was to
12:44isolate the C 71 positive cells and then
12:47to perform a Western blot with antibody
12:50specific for three nine, 298 on in.
12:53OK so here we see that only in C 71
12:57positive cells do we see an OK activation
13:00phosphorylation at this site but not in C 71.
13:05Get of cells and that's whether you have
13:07herpes 19 knocked down or control. Again,
13:10if we immunoprecipitated and OK in cells,
13:14either control or RPS 19 knockdown cells,
13:17we see that the phosphorylation in CD
13:2071 cells of these three substrates
13:23is is upregulated,
13:24and that's what we would expect because
13:28we get hyper activation of an OK.
13:31So this is only again in C 171671
13:34positive cells but not in other lineages.
13:40Furthermore, we looked at the
13:42activation of MLK in mouse models,
13:45so we collaborated with new on
13:48Flickr at Lund University in Sweden,
13:51who provided to us stem cells from RPS.
13:5519 knockdown mice.
13:56It's catcher cycling inducible.
13:58Similarly, we examine the stem cells from RP.
14:0211 flox mice, which is provided to us by
14:05Manuel Serrano from Barcelona, Spain.
14:08This is also.
14:10Tamoxifen induced to lose one allele
14:13of RP 11 gene and what we see is in
14:16these mice with which have anemia that
14:19the activity of in OK is much higher
14:23than our control mice or minus docs.
14:26In both the RPS 19 and RPL 11
14:29knockdown mice mouse cells.
14:31We also examined patient samples and
14:33this is a collaboration with Hannah
14:35Gosda from Boston Children's Hospital.
14:38She provided to us three
14:40different patient samples.
14:41With these mutations in our PS 19
14:44Mark then examined the relative in OK
14:48kindness activity compared to healthy
14:50control cells and show that that the
14:54activity was significantly increased.
14:56And finally,
14:57in collaboration with here Mitsunaga,
14:59which he had Stanford,
15:00his postdoctoral fellow,
15:02to she developed,
15:03I PS cells from two different patients
15:05who are diagnosed at Stanford,
15:07who had the typical clinical
15:09and physical features of DVA.
15:11And one of them had an RPS 26 mutation.
15:14The other one had an unknown mutation,
15:17but in all three clones that
15:19we analyzed from I PS cells,
15:21we see hyperactivation of an OK using
15:24the substrates that I mentioned.
15:26And OK leban rector.
15:31So given this in all of the
15:33models that we examined for DBA,
15:35regardless of the mutation,
15:36we were able to see an OK activation.
15:40And so we wanted to test the hypothesis
15:42could in OK be a possible target for therapy.
15:46The way that we looked at this was to
15:49study the number of two thirty 5235 cells,
15:52which is a reflection
15:53of birthright expansion.
15:54We see that in our PS 19 knockdown cells,
15:58if we treat with sin LK.
16:00To knock down MLK,
16:01we see increase in the numbers of these
16:04are three projector cells or three cells,
16:07in contrast to absence of in OK or.
16:10Design OK or a control escape in
16:12OK and the point here is that
16:15in the patients with DPA,
16:18generally speaking they are treated
16:20with steroids or transfused in
16:22when the hemoglobin is below 8.
16:24So our goal is not to necessarily improve
16:28the hemoglobin up to normal levels,
16:30but rather to increase or decreases
16:32sufficiently so that they will no longer
16:35need story therapy or wrestle transfusions.
16:38So that's really our goal.
16:42You have a hand raise. Pat Gallagher
16:45is raising his hand. OK, yes?
16:50Let me think we have to figure
16:52out how to let him talk.
16:54Hold on one second so I can do it.
16:56I can do it. Go ahead.
16:58I'm good guys are the question I got OK?
17:02Do you put my hand down there?
17:05I don't know if I can do know how to do that.
17:09I had a question though on the
17:11previous slide you normalized
17:12the healthy control cells.
17:14Yes to 100% on the previous slide.
17:16Yeah, and had to go back and if you
17:19knock down you had the sin LK and you
17:21gave the absolute number of cells.
17:24Is that affected in the healthy controls?
17:26Well certainly with DPA
17:27patient samples there are lower numbers
17:29of cells, but in spite of this we see
17:32hyperactivation of energy. So I'm talking
17:34bout in the controls on the left. In
17:37the controls, if we knock if we if
17:39we look at the numbers of cells.
17:42I don't know if we actually have done that.
17:45I mean normally we would try to
17:48normalize the cell's cell number.
17:51To make it equal on both sides,
17:53but yet it's still we see hyperactivation of.
17:57I mean we see lower levels of an OK
18:00activation. Yeah thanks yeah.
18:03So in collaboration with Johan,
18:05we were able to look at.
18:07He gave us 8 compounds that he had screened.
18:12Using a 14,000 compound library in his DBA
18:15RPS 19 knockdown mice trying to go back,
18:19but I can't. And what we saw was
18:23these eight compounds and compounds.
18:25Six and eight were the most effective
18:28as far as their third expansion.
18:30And then if we looked at the Enoch activity,
18:34the typical substrates that
18:35we've examined in the past,
18:37we see that in these two compounds we were
18:40able to see significantly decrease activity.
18:43Now here we have our Pierce
18:4519 knockout cells again,
18:46and we see with this is with sin LK.
18:50So when we add this compound is
18:52Scituate which is a TGF beta inhibitor.
18:55Which has an OK as an off target.
18:59We see that there's improvement
19:01in other places.
19:02However, if we knock down in OK,
19:04we still see an improvement.
19:06So just by knocking down and OK,
19:08we see an improvement as we would expect,
19:11and then the treatment also in combination
19:14with an saw an OK gave about equal
19:16amount of improvement of earth places.
19:19So what the suggested to us
19:21was that the effect of this SD,
19:23two ages most most likely
19:25due primarily to an OK.
19:27So pushing up in OK activity.
19:29There's no effect of Miley
19:30sales as we previously seen.
19:34This presentation also within OK
19:37Nebbish and improves with pre sis of
19:40of our mouse models again which we had
19:44shown but turn 19119 increased with
19:46treatment of our this TGF beta inhibitor.
19:50Same with RPL 11 in the fold was
19:53different a little bit but two to three
19:57fold two year 424 fold stimulation.
20:00A production of rich white cells.
20:05We we also looked at human.
20:07Our cell models, including our PSAT,
20:09knocked down and 11 knock down.
20:12We see again Approvement at our DBA cells.
20:15No ST208 with SD too late,
20:17so we see an improvement anywhere
20:20from four and a half to about 7
20:23fold and then finally a patient
20:25sample we because he 235 levels.
20:28And found where would CSD 208 that
20:30we see increase in our ourselves
20:33with red cells and above 2 fold
20:35increase by quantitating.
20:37So how does this occur?
20:39Well one of the reasons why we think
20:41MLK may be important for treatment such
20:44as losing is that Lucy and may require
20:48which is amino acid as I mentioned
20:50before in clinical trials that it
20:53may require active mtor complex and
20:55pathways to be activated or induced.
20:58And Raptor me and this whole complex.
21:01We believe this is attached to lysosome.
21:05In the case of an OK phosphorylating Raptor,
21:08we hypothesize that this released
21:10this complex from the license zone,
21:13thereby enabling translation to occur.
21:15So Mark perform evening up
21:17fluorescent experiments where he
21:19labeled Raptor with green life,
21:21some in red and then looked at
21:24colocalization in the case of
21:26RPS Anki knockdown cells where he
21:29saw was that the Raptor was cold.
21:31Localising with their lices own.
21:34And as attached seem to be attached
21:37or interacting with each other,
21:40whereas with an OK knock down we
21:42don't see wrapped or any longer
21:45colocalizes with their license home,
21:48so this is 1 possible role model that
21:51we propose for which YNLK inhibitors
21:54might actually work together with lysine
21:57to enhance the erythropoietic effect.
22:00And this is another experiment where
22:03we basically took our cell model RPS
22:0619 and looked at the number of two
22:09CD 235 cells and you can see here.
22:12So control losing increasing
22:14orbital Boyces St 208, again RTF,
22:16beta inhibitor and the combination of
22:19the two seemed to increase even more.
22:22The effects through synergistic and
22:24synergism and then the we looked at.
22:27Also for EBT one which is a downstream
22:30target of optimizing implore.
22:32And showed that the phosphorylation
22:34was also enhanced when we combine the
22:37leucine with SD in our PS90 knockdown models.
22:42So in summary of this,
22:44part of the talk we showed that in
22:47OK is activated in every projectors
22:49from DBA patient samples as well as
22:53our other human and mouse models.
22:55Pharmacological genetic inhibition of
22:57NLC increases with regenerx expansion
23:00in our models and then OK appears to
23:03exert influence on putting translation
23:04in DBA through the mtor pathway.
23:07So our focus was next to begin
23:10to identify potential therapies.
23:12That would target in OK.
23:14Unfortunately, the SD 208 compound is
23:17not ready for clinical application.
23:19It doesn't have the appropriate
23:22physical chemical properties,
23:23including solubility,
23:24to be able to be converted to the clinic.
23:28And So what we show is that we
23:31worked with a number of mid chemistry
23:34consultants from our Spark program,
23:37which is a program at Stanford to
23:40convert projects and lab to the clinics.
23:43And were able to examine a number
23:46of these compounds,
23:47all of which have different primary targets.
23:50But we tested a number of them.
23:52This is just a representative
23:54experiment showing in OK in
23:56vitro kinase activity descend,
23:58everyone is familiar with which
24:00is a tyrosine kinase inhibitor
24:02used to treat CML and carafe,
24:04and if it would be rough inhibitor OTS.
24:06167 is a milk inhibitor suppen
24:09assertive is an important hitter in
24:12St 208 is our TGF beta inhibitor.
24:14So just to show an example,
24:16we use low and high concentrations.
24:18We able to show significant decrease in OK
24:22activity with this particular drug OTS 167.
24:25We examined some of these compounds for their
24:28ability to increased risk of Oasis in RPS,
24:3119 lockdown models and here's
24:33our TGF beta inhibitor.
24:35We see various levels of
24:37increase in Rip Oasis,
24:39but mostly due to RTS 167.
24:42We also looked for phosphorylation
24:44of Raptor again,
24:45which is the target of which
24:47is in the Mentor complex and we
24:50see that there was an increase
24:52in phosphorylation of Raptor.
24:55But decrease with OTS 167 within OK activity.
25:02So our lead compound we decided
25:04to focus on was OTS 167.
25:07This is a milk inhibitor.
25:09It's currently under being studied
25:11in Phase 1/2 clinical trials and
25:13particularly in advance to acute
25:15leukemia as well as lung cancer.
25:17The drug was developed by uncle with
25:20therapy scientists and their bins.
25:22Number of studies that show that drug is
25:25still effective when milk is knocked out,
25:28so there are clearly other targets involved.
25:32This compound was shown in RP slinky
25:35knockdown cells again to increase
25:37every thread expansion and when
25:39we combine our knockdown oven OK
25:42with this inhibitor we did not see
25:44significant increase in irithyll.
25:46Police is suggesting to us again that
25:50the primary target of OTS in this
25:52system to improve it or places in OK
25:55and this drug was dosed at 200 animal
25:58or every three days for one cycle.
26:04To Sir to to understand whether
26:06there there there was toxicity
26:08to normal or through blast.
26:11We looked at every expansion
26:13as well as DBA cells.
26:15Knockdown cells with artist 19 we see
26:18that the maximum effect was at 300 animal,
26:21but we begin to see an effect in
26:24North places as early as 30 nanomolar,
26:27which shows greater than
26:29tenfold therapeutic window.
26:30Since the IC50 and normal.
26:32Or healthy with Glass was
26:34about 480 animal are.
26:36With my lead cells we saw slightly
26:39more sensitivity of this compound.
26:41Again, we can see 30 an animal or an
26:44increase in it with the police is,
26:47but then the myeloid cell we
26:49began to see decreased more
26:51significantly around 300 nanomolar,
26:53so Even so we believe the
26:55therapeutic window is about 10 fold.
26:57And again, this is an in vitro assay system.
27:02So in conclusion,
27:03Altius 167 appears to improve our
27:06ethical thesis in our DBA models in
27:09vitro with very little toxicity.
27:11There have been previous reports
27:13in Nora Blastoma and breast cancer
27:16Xenografted mouse models treated
27:17with OTS 167 twice a week for three
27:21weeks over the course of a month or so,
27:24and none of those mice developed
27:26bone marrow toxicity.
27:28There are also other inhibitors to
27:30indicate that potentially they.
27:32May be effective,
27:33including John Conyers inhibitors
27:35and we're currently testing those
27:38and finally experiments to really
27:40test this is necessary in vivo
27:42in order for us to proceed to
27:44clinical trial stage.
27:48Another interesting observation
27:49that Mark made as far as Enoch
27:52expression is the fact that metformin,
27:54the commonly used medication
27:56for type 2 diabetes,
27:58inhibits an OK expression in small
28:00cell lung cancer cells,
28:02and this drug also improves the
28:04effect of hematopoiesis and delays
28:07tumors in Fanconi mice.
28:08This has been reported to be and metformin
28:11to be protective against aldehydes,
28:14which is one of the toxins
28:17thought to affect inhibit.
28:18He met up with stem cells and
28:21he's in this disease.
28:22There is currently a phase two
28:24trial with metformin in Fanconi
28:26patients that's being directed by
28:29Akiko Shimamura Boston Children's.
28:31So what is the mechanism by which in Oak
28:33expression is inhibited by metformin?
28:35Well, one of the things we
28:38looked at was first of all,
28:40just metformin improve the cell
28:42numbers of C235 cells and we
28:44show that both in RPS 19 RP.
28:4611 Knocked down models that it
28:48does increase the production
28:49of these research senators.
28:51In contrast,
28:51there's no effect of moderate cells.
28:53We don't see any phenotype,
28:55and then it would be a few economy.
28:58Or if you eat colonies,
29:00we see an increased number.
29:02With metformin. In contrast to controls.
29:06CD 235 arthritis sales.
29:10Metformin increases by five or
29:12six fold two to six fold and
29:14then be a few mirrors for it's
29:16also about two or three fold.
29:18We've seen increasing without no
29:20effect on the mileage ourselves.
29:24The metformin also improves erythropoiesis
29:26by inhibiting anoche activity.
29:28Here we have again in OK
29:31phosphorylation of in OK Mabel Rector.
29:34In all cases we see that the metformin
29:38as well as SD 208 TGF beta inhibitor
29:43decreases the activity of a van.
29:46OK in our PS90 knockdown model.
29:49The RNA expression also decreases
29:51which is interesting which is.
29:54Primary mechanism by which we
29:57believe metformin inhibits an OK.
30:00Anorith Rd expansion we see
30:02again with knockdown of RPS 19.
30:04Any cells and knocked down again OK at
30:07the same time we see that just knock down
30:10of in OK and improves mini expansion.
30:13There is replaces but also just say it not
30:16kind of in OK here as well as metformin.
30:19So metformin not done OK in the
30:21form it again no significant
30:23change which again suggested some
30:25informants working through an OK.
30:30Sorry.
30:33This is a slide which shows our
30:36treatment of zebrafish models in
30:38collaboration with Scholin at UCLA.
30:41He created a DBA model
30:43using RPS 19 morpholinos.
30:45This is a phenotype of the fish.
30:48After approximately 5 days you can see
30:51that these embryos show significant
30:53anemia compared to controls with
30:55the treatment with metformin.
30:57We see that there is again increase
31:00in risk reduction as indicated
31:02by staining with Odeon sitting.
31:05Which binds to hemoglobin.
31:09So how does metformin
31:11regulate in OK expression?
31:13One of the ideas is through micro RNAs,
31:17and so Mark created a number of
31:20truncation mutants that would include
31:22include a variety of micro ironies
31:25for which we can then try to identify
31:29the specific mechanism by which
31:31metformin effects in OK expression.
31:35So this is showing the levels in
31:38humans that mirror 30 a mere 26
31:42increases with metformin treatment.
31:44So this is again in human primary
31:47cells versus mice that do not show
31:51this increase in May 26 expression.
31:54So the idea here is metformin could be
31:57inducing near 26 which binds to the
32:01three prime UTR event OK resulting
32:04in inhibition of by of expression.
32:07So this is a luciferase constant
32:10Reporter construct showing activity as
32:12a reflection of transcription of in OK.
32:14You can see that with the R in micro
32:17RNA 26 mimetic that there is we see
32:21decreased transcription also with
32:23181 which is in the same region
32:26the truncation mutant as mere 26
32:28we see a decrease in me.
32:31Yeah can you
32:32clarify what you mean by a mere 26
32:35mimetic? It's it's a similar.
32:37The structure, as a mere 26,
32:39so it's acting as if it were a mere
32:4226 and binding to that site, so I hope
32:45it's done with like with the
32:47retrovirus like with a hairpin or.
32:48Yes, I believe so, yeah.
32:51And the same thing with an OK if
32:54we we see that in OK expression
32:57at the protein level is decreased.
33:00But a sponge which basically is
33:02exactly what it described it in
33:05no longer allows me or 26 to bind
33:07to the three point Muti UTR.
33:10We see that there is again
33:12increase in OK at the Mr. NY.
33:14Also prefer at the Reporter assay
33:17and also the protein level.
33:20And also the Mirror 181 on the top
33:23didn't affect, but on the bottom did
33:25what's going on with that?
33:27Let me see for here.
33:29Yeah, I mean that's a good question whether
33:32the mere 181 it may not be a specific
33:36so that it doesn't sound consistent,
33:38because you would expect me
33:40or 181 to increase there too.
33:42So you know, not all the mirror when
33:45he I guess they could be inhibiting
33:48or blocking other mere sites.
33:50So, but that's something we
33:52need to look at more carefully,
33:53but thanks for noticing, yeah.
33:56An Emmy this metformin also
33:58mediates everything voices,
33:59so again we see the C235 increase
34:02with the mimetic that Foreman or both.
34:06Finding interrupted right
34:07'cause they're happening.
34:08So there's a question from the audience.
34:11Is the difference of metformin
34:12effect on human versus mouse
34:14near 26 due to changes in mere
34:1726 transcription or processing?
34:18Yeah, that's a really good question.
34:21I don't think we really know that.
34:23We haven't really focused on the
34:25mouse system, but we do know that it's
34:28different and you know that there are
34:31clearly differences at the genomic level
34:33and that the exact reason for that.
34:36We're not sure. So it's a good.
34:39It's a great question, yeah?
34:42So both in the mouses are
34:45in the mouse system.
34:47We see an increase turn 119 and also 235.
34:51We see increase in CD 235 R 3119
34:55the expression as you would expect
34:58goes down with metformin treatment
35:00here in our PS90 knockdown cells.
35:04So, to summarize,
35:05metformin improves with crisis
35:06in our models of DPA.
35:08It decreases in OK expression through
35:10mere 26 A and targeting mirrors is
35:13a possible approach to DBA therapy.
35:15There now companies that are
35:17trying to make medics or sponges
35:19for clinical application.
35:20Although it's very in early stages.
35:25OK, for the last part of the talk I'm
35:28going to focus on a new project that I
35:32appreciate Diane and Vanessa's input.
35:35This is a protein set B1 which is
35:38special 80 rich binding protein one
35:40and we initially did a many years ago.
35:44Actually did a RNA seek experiment
35:47with fetal liver human CD 34 positive
35:50cells that were transduced with RPS 19.
35:54Like if I were constructs and found 560
35:56or so genes that are differentially
35:58expressed and then we cross reference that
36:01with the list of genes identifying earlier.
36:04It'll poesis in a paper published by
36:07Mohan Orla and also Pat Gallagher and
36:09found about 1700 genes over lapping were
36:12about 42 genes that we then analyze
36:15in a variety of our cell model systems
36:18and hematopoietic stem cells both in
36:20control an RPS 19 knockdown cells
36:22and found that among those that were.
36:25Mostly regulated,
36:26the Sepy one was very interesting to
36:29us since not much had been described
36:31at all on the role of sappy one during
36:35Aritha Poesis So sappy one is a
36:38protein that basically forms chromatin
36:41loops and regulates transcription.
36:44And there's a number of their number
36:46of papers that have very nicely
36:49described its expression and rolling
36:51hematopoietic differentiation.
36:52It's moderately expressed in HS
36:54season is required for self renewal.
36:57It's induced in lymphopoiesis and
36:59required for T cell expansion.
37:01Knockout mouse have defects in
37:03lymphopoiesis and then downregulation
37:05in Milo.
37:06Police has been demonstrated in its
37:09requirement for PU .1 regulation in
37:12in common Milo progenitor cells.
37:14So Mark looked at sappi 1M RNA
37:17expression anarchist.
37:1819 knockdown cells,
37:19day OD five.
37:21He showed that the expression
37:23decrease more rapidly in our PS 19
37:26knockdown cells compared to controls
37:28and then also at the protein level.
37:31We see that there is much more
37:33decrease in the protein levels.
37:36Is that being one day 5?
37:40The colony essays and liquid culture
37:42essays that we perform showed
37:45that in colony essays in RPS.
37:4719 knockdown.
37:48HSBC's that we did not see significant
37:51increase in colony numbers per
37:53plate with three expressions.
37:55We expressed that being one about
37:57to fold in these knockdown cells,
38:00we can see any much increase in
38:03liquid culture. However, we sell a CD.
38:06235 sales that there was increase in.
38:09With happy one overexpression.
38:12But although there was no increase
38:15in the colony numbers,
38:17we could see by visualization that on
38:19the colonies were huge for much larger,
38:22suggesting that there was a
38:24proliferation of these cells,
38:26and we did see also a normal controls,
38:29but not as dramatically as in our case,
38:3219 knockdown cells.
38:34There's also a hand up from. Yeah, it's
38:37Vince. OK, I've is.
38:42Sorry, that was sorry that
38:44was accidentally did that.
38:46Oh alright, never mind.
38:47So colony assays we see that in
38:50knockdown of sappy when in normal
38:53cells or healthy progenitors we see a
38:56slight decrease in an doxycycline or 71
38:59knockdown was 235 in liquid culture.
39:01So we do see a mild decrease in
39:05erythroid colonies and or through glass.
39:08So sad he went on.
39:10Regulation does not dramatically
39:11impact anemia phenotype in DBA.
39:13So what is the function of 71?
39:16So healthy uncommitted Mylar
39:18projectors we see that.
39:19So again,
39:20here's our colony assay results
39:23showing mild decrease in be a few.
39:25Ease the culture.
39:27The same 235 cells but in megakaryocytes
39:30regenerators we see at least a 30%
39:33decrease in with that be one knockdown.
39:36So our hypothesis at this point
39:38was that although in DPA we see
39:41a block here early erythroblast
39:44stage or seek to the 235 stage.
39:47That perhaps happy one is
39:49acting upstream of this,
39:51affecting the megakaryocyte
39:53recite progenitor stage.
39:55And so here we have again
39:57our as our PS90 knockdown.
39:59Showing calling or calling numbers.
40:01Interesting Lee with CFU GM.
40:03We see an increase in colonies.
40:07C235 expansion was noted with RPS 19.
40:10Knock down an 71.
40:14We did see also 11B expansion
40:17and overexpression.
40:18We also see CD 41 increase
40:21more even more significantly.
40:25How does this happen then?
40:27What is the mechanism downstream of that?
40:30Be? One well, one of the things one of
40:33the genes that has been very important
40:36recently has been a heat shock.
40:3970th proteins that's encoded by
40:41the gene zedex, HSP, 1A1B, and 1A,
40:43and there are three different papers.
40:46Sorry, there are three different
40:48papers that have been describing
40:50by Hermione and Leah Dacosta.
40:51The role of HSP 70 in regulating gotta one.
40:56An agency 70 is thought to interact
40:58with God and one to prevent the caspase
41:01dependent cleavage of God in one,
41:03so basically is stabilizing that
41:05complex or God or one to allow
41:08it to to to activate genes,
41:10and this is just a profile of chromosomes
41:13looking at various genes so they are not
41:16affected by sappy one or influence to
41:18rescued by setting one in the orange.
41:21So here are the two genes and if we
41:24knock down step one, we see that.
41:27This HP A1A expression goes down.
41:30Similarly,
41:30if we knock down at SABI one
41:33we see HSP one being Arnie,
41:36decreasing ansim with the protein levels.
41:41If we then look examine our
41:43peacemaking knockdown cells and
41:45then overexpress sappy one,
41:47we see that there's a rescue and
41:49the expression of these two genes
41:52and also at the protein level here.
41:58To understand what sappy one might be doing,
42:02and regulating house regulating it,
42:05just P1A1A and the HPA 1B.
42:08Mark performed chromatin
42:09immunoprecipitation assays with antibodies
42:12that are specific to these H3K914,
42:14desolation HK4 trimethylation
42:16HCC 27 five metalation,
42:18and these are the sites that set the
42:23one binds and what we see is that.
42:27There's also commented peaks which that
42:29with the H3K29 acetylation and HGK
42:324 trimethylation but not with H3K27
42:35Trimethylation. So here's our model.
42:37Here is the HSP A1A gene. Here's.
42:41Here are the seven one binding sites,
42:44and here are the predicted enhancer and
42:47promoter sites that we would expect,
42:50and so sappy one promotes the looping
42:53so that all of these will be enclosed.
42:57Doximity to facilitate transcription
42:59of these two genes,
43:00another approach that Mark Hughes
43:03was as chromatin confirmation capture
43:05essay where you basically treat
43:07the cells with formalin,
43:09then digest using restriction enzymes
43:11and then perform PCR and what he
43:15showed was that with our with our
43:18control we see two nice peaks at the
43:21sites of the Enhancement Promoter
43:23regions with RPS 19 knock down,
43:25this is blunted.
43:26With knock down as happy when we
43:28basically don't see these peaks and
43:31then with re expression of happy
43:33when we see rescue and again the
43:35peaks appear again.
43:37Another way to validate these data
43:40and look at it functionally is to
43:43perform a new technology known as Cloud Nine,
43:46and this is a approach that
43:49was developed by Kevin Wang at
43:51Stanford where you use Casper,
43:53Chris crisper CAS 9 to basically
43:56target certain sequences in the genome,
43:58treat with abscisic acid,
44:00which then forms and reversibly
44:02loop that's induced,
44:04and then you can see the effects downstream.
44:08So this is what Mark did and this
44:10is again showing our sappy one,
44:12and the various promoters
44:13and enhancer region,
44:14and again using G Arnie dimer pairs.
44:17If you knock down S isep or sappy
44:19one treat with this, I said the one,
44:22you see that there's no expression
44:24of these two.
44:25He drop protein genes.
44:27If we then induce looping using
44:30Cloud nine at at the EP1,
44:32which is here to transducer to
44:35facilitate expression of his HSP.
44:37A1A we see that there's expression
44:40but not with the other JPA 1D.
44:43If you then loop at EP2,
44:46which is here so that the enhancer
44:49is next to the P2 for HSV one a B1B,
44:53you see that there's expression of
44:56this gene and also slightly of HPA 1A.
45:00And then finally this is just a
45:03control showing there's no induction Eugene.
45:06So correlating with that is also
45:09me P expansion.
45:10We we knocked down SFB one,
45:13but then we expressed it.
45:15We see increase.
45:16We also see it by looping of P1
45:20and P2 increase in comparison
45:22to controls and we don't see any
45:25effect on HCS or the CMP population.
45:29So can
45:29I just clarify what's the experimental
45:32design here for this expansion
45:34and the CMP right? Starting with
45:37this is looking at 34 positive CD,
45:4071 low population of cells,
45:42but they've been cultured in vitro
45:44plus and minus for how yes, all of these
45:47yeah and then treated with the cloud 9,
45:50so you're then treating with the
45:53dimer peers and then with abscisic
45:55acid to induce the looping. But
45:57what you're starting with me P.
46:01We're starting at the beginning and
46:03then waiting until the stage where
46:05we would see these cells appear.
46:07OK, yeah. Yeah, and then we also look.
46:12We also examine the Seppi 1S I or any
46:16induced Rizzo Blastic Buster Apolysis
46:18showed that with the looping at P1
46:22and P2 we see increasing with robust
46:24increase in CD41A positive cells
46:27but not in Salem be my light cells.
46:32The similar findings were also
46:34observed in cells that had SHRPS 19.
46:36So in our PS90 knockdown cells
46:38the same thing happened.
46:40If we loop it P1 we see
46:43the expression of HSP 1A.
46:45If we loop at P2P CHSP A1B
46:47expression and then controls,
46:49we don't see and then this is just
46:52Western blot showing the expression
46:54a day five for both of these.
46:58Experiments.
47:01So looping appears.
47:02He also rescues Happy one,
47:04defects and healthy megakaryocytes.
47:07Here's colony essays showing
47:09our looping experiments and
47:11knockdown of 71 so by looping
47:14itself we can see increasing BFUE.
47:17Nothing with the CFU GM.
47:19Also, in liquid culture we see
47:21just with looping alone we can see
47:24improvement and C 235 and 41 a,
47:26but not in CD11B.
47:27So again, that tells us that
47:30the looping is critical for the
47:32transcription of these genes to
47:34take place and also for the effects
47:38on CD235 and C41 proliferation.
47:40Finally, to link this all back to God.
47:43Oh one. As I mentioned earlier,
47:45the God one is regulated HSP 70 being
47:48available to prevent the degradation of God.
47:51Oh one.
47:52So here we see 235 positive
47:54cells E 41 positive cells,
47:56gotta one expression.
47:57And if we then take our knockdown
48:00of essay of sappy one,
48:01we see that the looping can restore
48:04the expression of God in one because
48:07now you're getting HP 70 expression.
48:09So it's able to prevent
48:11degradation of God 01.
48:13Godwin expression is just shown
48:15here showing the increase in
48:17expression correlating with with
48:19the data I just talked about in
48:22an MVP expansion also correlated
48:24with with the God one expression.
48:26So if you did a western on
48:29the top right for the HSP 70,
48:32it would parallel the gotta one.
48:34That's what we believe. Yes, yes.
48:36And so that goes back to our model where we
48:40are in normal conditions that 71 gradually
48:43decreases as cells become more committed.
48:47But we do see HP 70 increasing, which
48:50prevents gotten one from being degraded.
48:53There's some city state level there,
48:56but obviously once committed
48:57to pull through blast stage,
49:00we see significant increase in God,
49:02one in order to transduced jeans that are
49:05required for earthway differentiation,
49:08terminal differentiation.
49:09We knocked down sappy one prematurely,
49:12or, in the case of RPS,
49:1419, insufficiency, seven.
49:15That would be 1 levels are much lower,
49:18which does not allow HSP 70 P.
49:21It just be 70 should be expressed,
49:24which then blunts the expression of God.
49:26Oh, one so that we don't get the
49:29normal increase and got a wine.
49:32And we believe this contributes
49:34to the anemia phenotype.
49:36It's just a schematic of what we
49:39believe is happening in CMP, ME P.
49:42The ratio relative ratios that
49:43occur at this stage normally,
49:46and then at the risible stage
49:48would receive an abundance of God.
49:50One protein that basically
49:52overwhelms the HSP 70,
49:53but that we have enough to be able
49:56to induce Raceway specific genes,
49:58so I'd like to end there.
50:01And then just to acknowledge
50:02people who have done the work.
50:03I think I've been knowledge Mark
50:05throughout the talk as well
50:06as the other Members I lab who
50:08contributed to this project.
50:09The collaborators at Stanford
50:10and also the other collaborators
50:12that not at Stanford in,
50:14particularly Diana Vanessa,
50:15for their help in doing the experiments
50:17with happy One and Megakaryocytic with
50:19recite progenitors and the funding
50:21sources that supported this work.
50:23So I'm happy to answer questions.